OSLC-OP PGB meeting Tuesday, 21 March at 8:00 DST, meeting notes

Agenda:

1. Status on OSLC Specifications (TRS 3.0 COS, Configuration Management 1.0 COS, Change Management errata)
2. Status of Linking Profile 
3. Status of Link Discovery Management spec
4. Proposal for OSLC Product Workgroup

Spec updates

TRS: v3.0 PS02 has been published. This revision contains updates requested by OASIS It is ready to promote to COS and go to public review.

Action, Jim: Schedule the SMV COS vote

Config 1.0: v1.0 PS01 has been published. This Product Specification is ready for COS pending SOU from IBM.

Action, Jim: Follow-up with IBM OASIS Representative

Change Management v3.0, OS: Errata completed to restore missing QM link properties needed when using configuration management.

Action, Jad, David: Complete pull request review

Action, Jim: merge approved pull request and work with OASIS on how to deliver the updated documents.

Linking Profile

OSLC enables a lot of required server implementation flexibility with MAYs and SHOULDs.

This flexibility is essential for server implementations to meet their specific needs.

But it leads to variability that inhibits interoperability between servers and increases development complexity.

The Linking Profile specification was intended to manage that variability somewhat to foster integration by providing different profile conformance levels through tightening MAYs and SHOULDs to MUSTs.

A draft specification has been created, and the scope of work and profile compliance points are known.

We have had some difficulty making progress due to limited contributions/participation.

Should Linking Profile be an OASIS spec or a Project Note providing implementation guidance?

Andrew originally proposed it should be a spec, but use different profiles to handle the variability.

The intention was for customers to be able to determine what integrations they need and to know what profiles their chosen tools would need to provide.

Customers are interested in the profiles, but perhaps vendors are less motivated because of the development effort to achieve conformance.

Are there any publicly available details on Polarion and Jazz integration?

No, these are private. But authentication was a significant issue.

Sodius-Willert and MID also have a lot of experience with OSLC integrations.

Consensus was to proceed on the standards track for the Linking Profile, but consider publishing as a project note once we have enough content to assess the pros and cons.

Link Discovery Management (LDM) Spec

An initial LDM 1.0 WD01 draft has been created, with an abstract and introduction.

Eran and Martin are editors.

The scope is reasonably well understood, except for possibly how an LDM server summarizes the scope of links it provides.

The PGB recommends creating a model of LDM clients and servers, and their resources, with sequence diagrams that describe the interactions for link discovery.

Then develop the REST APIs needed to support the data model and sequence diagrams.

Michael Rowe will help with LDM REST API details.

OSLC Product Workgroup Proposal

Gray presented meeting notes from Product Workgroup meeting that summarizes problem to be solved and value proposition.

Erik Herzog says: I think we need to consider two use cases: 1. Traditional OSLC to Tradtional PLM. 2. Traditional PLM to Traditional PLM.

Should we also explore ProStep and SysML v2 to understand what role they might play in a broader systems of systems integration and lifecycle management problem?

What are the overlaps with these standards and PLM?

We have understood the ALM/PLM integration problem for some number of years, but have struggled to make progress and deliver real integration value, despite specific customer needs.

We may need to understand these challenges to better inform the problem we are trying to solve.

Open discussion on possible future work or direction?

No further discussion.